Selected article for: "important feature and random forest"

Author: Xavier Hernandez-Alias; Martin Schaefer; Luis Serrano
Title: Translational adaptation of human viruses to the tissues they infect
  • Document date: 2020_4_7
  • ID: 0rk2dw4e_14
    Snippet: In an attempt to understand which tissues are the most predictive in identifying the viral tropism of proteins, we analyzed the relative feature importance within each random forest classifier, which measures the contribution of each tissue SDA in the decision trees (Fig. 2B) . The main observation is that no single tissue alone is able to discriminate against the specific tropism, since all feature importances lie below 0.10. However, it is also.....
    Document: In an attempt to understand which tissues are the most predictive in identifying the viral tropism of proteins, we analyzed the relative feature importance within each random forest classifier, which measures the contribution of each tissue SDA in the decision trees (Fig. 2B) . The main observation is that no single tissue alone is able to discriminate against the specific tropism, since all feature importances lie below 0.10. However, it is also clear that translational . CC-BY 4.0 International license author/funder. It is made available under a The copyright holder for this preprint (which was not peer-reviewed) is the . https://doi.org/10.1101/2020.04.06.027557 doi: bioRxiv preprint adaptation to bile duct (CHOL, for healthy samples of cholangiocarcinoma) is a recurrent discriminant feature, while other tissues are specifically important for just one or few tropisms, such as rectum (READ, for healthy samples of rectum adenocarcinoma) in predicting intestinal viruses. In any case, the directionality of these features cannot be established.

    Search related documents:
    Co phrase search for related documents
    • decision tree and feature importance: 1, 2, 3, 4, 5, 6, 7, 8, 9
    • decision tree and forest classifier: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13
    • decision tree and random forest classifier: 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13